Brain disease diagnosis service apparatus and brain disease diagnosis service method

Abstract

The present disclosure relates to a brain disease diagnosis service apparatus including at least: a receiving unit for receiving an image of a subject to be diagnosed; a preprocessing unit for estimating a transformation matrix in order to align a predetermined standard image with the image received by the receiving unit, and for applying the estimated transformation matrix to a predetermined standard brain region map, thereby generating an individual brain region map for the subject to be diagnosed; a feature point extracting unit for calculating a ratio occupied by brain lesions for each brain region in the individual brain region map, generated by the preprocessing unit, of the subject to be diagnosed, thereby extracting feature points; and a disability type determining unit for determining a disability type on the basis of the ratio of the brain lesions, calculated by the feature point extracting unit, for each brain region.

The invention claimed is: 
     
       1. An apparatus for servicing a brain disease diagnosis based on an image obtained from a magnetic resonance imaging (MRI) apparatus, the apparatus comprising a processor, wherein said processor comprises:
 a receiving unit configured to receive an image of a diagnostic subject obtained from the MRI apparatus; 
 a preprocessing unit configured to
 estimate a transformation matrix for matching a preset standard image to the image received from the receiving unit and 
 apply the estimated transformation matrix to a preset standard brain region map such that an individual brain region map of the diagnostic subject is generated; 
 
 a feature extracting unit configured to calculate an occupancy ratio of a brain lesion of each brain region in the generated individual brain region map of the diagnostic subject 
 a disability type determining unit configured to determine a disability type based on the calculated occupancy ratio of the brain lesion of each brain region; 
 a disability degree classifying unit configured to classify a disability degree corresponding to the calculated occupancy ratio of the brain lesion of the diagnostic subject, based on a reference value set by using occupancy ratios of brain lesions of diagnostic subjects having the same disability type as that determined by the disability type determining unit and 
 a prognosis predicting unit configured to predict a prognosis of the determined disability type in accordance with the classified disability degree. 
 
     
     
       2. The apparatus of  claim 1 , wherein the image of the diagnostic subject obtained from the Mill apparatus comprises apparent diffusion coherent (ADC), diffusion weighted image (DWI) and gradient-echo (GRE) images. 
     
     
       3. The apparatus of  claim 2 , wherein the preprocessing unit is configured to match the ADC and DWI images to each other to extract a position of the brain lesion. 
     
     
       4. The apparatus of  claim 2 , wherein
 the standard image comprises a standard T2 image, and 
 the preprocessing unit is configured to
 estimate an affine transformation matrix to match the standard T2 image to the GRE image and 
 generate standard T2 image converted in a brain shape of the diagnostic subject through the estimated affine transformation matrix. 
 
 
     
     
       5. The apparatus of  claim 1 , wherein the disability type determining unit is configured to
 generate a classifier, by using disability types and occupancy ratios of brain lesions of diagnostic subjects, to determine a reference value of a brain lesion in each brain region, and 
 determine the disability type based on a position of a brain region exceeding the reference value determined by the classifier and the calculated occupancy ratio of the brain lesion. 
 
     
     
       6. The apparatus of  claim 1 , further comprising
 a storage unit, 
 wherein the prognosis predicting unit is configured to provide the prognosis predicted by analyzing treatment results of diagnostic subjects who have the same disability type and occupancy ratios of brain lesions in a predetermined range with reference to brain lesion positions, the occupancy ratios of the brain lesions, disability types and the treatment results of the diagnostic subjects stored in the storage unit. 
 
     
     
       7. A method of servicing a brain disease diagnosis based on an image obtained from a magnetic resonance imaging (MRI) apparatus, the method comprising:
 receiving an image of a diagnostic subject obtained from the MRI apparatus; 
 estimating a transformation matrix for matching a preset standard image to the received image and applying the estimated transformation matrix to a preset standard brain region map such that an individual brain region map of the diagnostic subject is generated; 
 calculating an occupancy ratio of a brain lesion of each brain region in the generated individual brain region map of the diagnostic subject to extract a feature by; and 
 determining a disability type based on the occupancy ratio of the brain lesion of each brain region; 
 classifying a disability degree corresponding to the calculated occupancy ratio of the brain lesion of the diagnostic subject based on a reference value set by using occupancy ratios of brain lesions of diagnostic subjects having the same disability type as that determined; and 
 predicting a prognosis of the determined disability type in accordance with the classified disability degree. 
 
     
     
       8. The method of  claim 7 , wherein the image of the diagnostic subject obtained from the MRI apparatus comprises apparent diffusion coherent (ADC), diffusion weighted image (DWI) and gradient-echo (GRE) images, and the generating of the individual brain region map comprises matching the ADC and DWI images to each other to extract a position of the brain lesion. 
     
     
       9. The method of  claim 8 , wherein
 the standard image comprises a standard T2 image, and 
 the generating of the individual brain region map comprises
 estimating an affine transformation matrix to match the standard T2 image to the GRE image, and 
 generating a standard T2 image converted into a brain shape of the diagnostic subject through the estimated affine transformation matrix. 
 
 
     
     
       10. The method of  claim 7 , wherein the determining of the disability type comprises
 generating a classifier, by using disability types and occupancy ratios of brain lesions of diagnostic subjects, to determine a reference value of a brain lesion in each brain region, and 
 determining the disability type based on a position of a brain region of a brain lesion exceeding the reference value determined by the classifier and the calculated occupancy ratio of a brain lesion.